[00:09] robert_ancell, what are your thoughts on using libgnome-desktop3 to set the average background color? I'm not sure I want to propose rewriting how we do the background stuff we have at this late juncture, so it would mean we'd be using its background setting method even though the user never saw the root window [00:09] robert_ancell, this is in reference to bug 1043572 [00:09] Launchpad bug 1043572 in Unity Greeter "Shade notification colors according to current background" [Undecided,New] https://launchpad.net/bugs/1043572 [00:10] mterry, should be ok [00:11] robert_ancell, I'll try to whip something up, but in order to make it for UIF, I'll try to finish it tonight so you can review [00:11] ok === nijaba_ is now known as nijaba === cpg is now known as cpg|away === cpg|away is now known as cpg [04:35] Good morning === sgnb` is now known as sgnb [06:24] how do you create animated progressbars in gtk-3 ? === sagaci_ is now known as sagaci [07:00] good morning === axp2_ is now known as axp2 === axp2_ is now known as axp2 === henrix_ is now known as henrix [09:06] wtf [09:06] my e-mail to ubuntu-devel has been moderated [09:06] "The reason it is being held: [09:06] Post by non-developer to moderated list." [09:06] Ugh. I'm the owner of ubuntu-core-dev. Pretty sure I'm a developer ;p [09:07] cody-somerville, sent under wrong address ? [09:07] Checked [09:07] cody-somerville: wrong email alias? which email are you subscribed to with to the mailing list? @ubuntu? [09:07] Sender: cody-somerville@ubuntu.com [09:08] did you stop developing and mailman noticed ? :) [09:08] * xnox ..... who will break the news out to cody?! =))))))))))) *giggles* === tkamppeter_ is now known as tkamppeter [09:08] possibly :-) [09:08] mailman is written by barry so it's probably got all sorts of features ;) [09:09] haha === ev_ is now known as ev === morphis_ is now known as morphis === cpg is now known as cpg|away === mpt_ is now known as mpt === yofel_ is now known as yofel === Pendulum_ is now known as Pendulum === _salem is now known as salem_ [12:37] can someone help me understand why 'apt-get install bacula' in 12.04 results in 'bacula : Depends: bacula-server but it is not going to be installed' [12:37] stgraber: do you know to whom i would speak to learn how to set up ubuntu studio's profile for autotesting per https://blueprints.launchpad.net/ubuntu/+spec/other-q-ubuntu-flavors ? [12:38] I figure its something todo with unversioned dependencies on the bacula meta-package but I'd like to understand why it creates this issue (if that is the problem) [12:38] point me at documentation, an example, or another person would be helpful :) [12:41] jamespage: good question and i dont think you are being a dumbass [12:47] jamespage: tested in my precise vm and apt is happy with the dependencies. try if "apt-get install bacula bacula-server" tells you more (you might need to repeat this with additional dependencies till you get to the real error) [12:48] geser, hmm - "apt-get install bacula bacula-server" works just fine [12:48] interesting [12:49] geser, there is a version of bacula in -updates as well which I think is causing the problem [12:49] bacula -> bacula-server [12:49] bacula does not have a versioned Depends on bacula-server [12:50] I've -updates enabled in my precise vm and "sudo apt-get -s install bacula" shows no error [12:52] geser, odd - I get http://paste.ubuntu.com/1175807/ from exactly the same command [12:53] is this about an upgrade or fresh install of bacula? [12:54] geser, fresh install [12:54] relates to bug 1034962 [12:54] Launchpad bug 1034962 in bacula (Ubuntu) "bacula dependency on bacula-server fails with held broken packages" [Undecided,Confirmed] https://launchpad.net/bugs/1034962 [12:54] geser, my instance is fully up-to-date and is amd64 arch [13:02] jamespage: did you try if temporarily disabling the -updates repo "fixes" it? [13:03] geser, good point [13:03] * jamespage tries it [13:05] hmm - odd - I still see the same issue [13:05] * jamespage puts the idea about -updates confusing things in the bin [13:10] try activating some of apt's Debug settings like Debug::pkgProblemResolver, perhaps it gives a hint [13:34] geser, thanks for that pointer - I think I see the issue [13:35] I think its a Recommends down the depencency chain on a virtual package - this is fixed in quantal [13:36] precise: Recommends: bacula-sd-tools [13:36] quantal: Recommends: bacula-sd-mysql (>= 5.2.6+dfsg-2ubuntu1) | bacula-sd-tools [13:37] jodh: SpamapS: the upstart cookbook doesn't seem to mention anything about 'restart job' acting different from 'stop; start' when there is a pre-stop or post-stop stanza. that is the case right? [13:39] zul, geser: its broken in oneiric as well [13:39] * jamespage sighs [13:41] jamespage: you're probably right. I had Recommends disabled my precise vm and with Recommends enabled I get that error too [13:42] geser, testing that theory now; the Recommends has been like that since forever so suspect a change in resolver behaviour as well - but its def not right [13:42] but 'sudo apt-get -s install bacula -oApt::Install-Recommends=false' works again [13:42] geser, thanks for the help BTW [13:50] gema, ^^ why we need more automated test coverage of seeded packages which are not on ISO's [13:51] jamespage: excuse my ignorance, where do I find those seeds? [13:52] gema, launchpad - they are all owned by the ubuntu-core-dev team [13:52] https://code.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/platform.quantal for example [13:52] jamespage: the interesting question is if this is a packaging bug in bacula or a regression in apt [13:53] geser, depends on what apt should do when trying to resolve a virtual package === tremolux_ is now known as tremolux [13:53] I find it odd that it just picks one [13:53] hallyn: well spotted. Yes, that is the case - the behaviour needs documenting in the man pages too. I'll add to the Cookbook this week if I get a chance. [13:54] jodh: thanks, wanted to make sure [13:54] mvo, any opinion on the discussion above? [13:55] assuming it ever worked as expected [13:55] geser, quite [13:56] btw this was Debian bug #602222 [13:56] Debian bug 602222 in bacula "bacula meta-package not installable with recommended packages" [Serious,Fixed] http://bugs.debian.org/602222 [13:59] geser, looks like thats been dropped during merges for some time.... [14:00] geser, there is certainly a behavioural change in apt between natty and oneiric which reveals this issue [14:08] hallyn: I don't know if it answers your question, but restart and stop;start differ in that stop;start stops the job as upstart knows it in-memory and then starts it anew based on what's in the upstart job file now. Restart doesn't care what's on disk. It uses it's in-memory copy of the job description as it were when it was started. [14:08] hallyn: Whether there are differences wrt to pre-stop and post-stop handling.. I don't know. [14:12] gema, dpm, ScottK, bdrung: ready? :) [14:13] dholbach, in 45 mins? [14:13] dholbach: dpm is starting, I am second! [14:13] yep [14:13] ready [14:13] just asking all speakers of today [14:13] gema, yes dholbach is aware too, we've changed it in the calendar. Thanks again for swapping :) [14:14] dpm: no problem! [14:14] dholbach: yes as a ill developer can be [14:14] soren: yea, that bit is documented in the cookbook. but also if there is a pre-stop or post-stop stanza, then 'restart' does not kill and restart it. i only know bc that's been a problem with libvirt before [14:14] biab [14:19] just to report something that ubuntu devs should read http://www.wired.com/wiredenterprise/2012/08/osx-killed-linux/ that's maybe why developers can't do a good programs for years things are keeping changing and they broke always the code you did before [14:27] Hello. If a debian bug targets the wrong package, is there a way to change the package it was targeted against, or should I file a new bug? [14:27] arges: http://www.debian.org/Bugs/server-control [14:27] pitti, thanks [14:28] arges: if you have the "bts" tool working, "bts reassign 12345 new-package new-version" [14:28] otherwise mail the command to control@bugs.debian.org [14:28] ahh very nice === jussi01 is now known as jussi [14:33] hello, could an archive-admin look into accepting clutter-gst-2.0 ? -- https://bugs.launchpad.net/ubuntu/+source/clutter-gst/+bug/1040930 [14:33] Launchpad bug 1040930 in clutter-gst (Ubuntu) "FFe: Sync clutter-gst-2.0 1.9.90-1 (universe) from Debian experimental (main)" [Wishlist,Fix committed] [14:36] ev, is e.u.c broken? [14:36] seb128: working fine here [14:36] ev, selecting: 12.10, gnome-control-center, the past month gives me a "No data to display" which I find hard to believe ;-) [14:36] like I know g-c-c has bugs, it should have collected at least one in august [14:37] there is a solid list of those for 12.04 though [14:38] seb128: it's a sort of bug in the deployed version of errors.ubuntu.com.... [14:39] it only shows problems with instance counts >= 20 [14:39] whereas trunk takes the 100 problems with the highest instance counts [14:39] so no g-c-c bug got >= 20 during august [14:39] that's hard to believe [14:39] so looking at trunk, I see problems with 18, 16, and 9 instances (and so on) [14:39] in quantal. [14:40] ok, maybe we have less quantal users that I though [14:40] some bugs have > 3000 in precise for g-c-c during august and they are not fixed in quantal [14:40] in 12.04 we had 3447 instances for the biggest g-c-c problem [14:40] yeah, I'm just surprise that we don't have a magnitude of some hundred for quantal at least [14:41] * xnox quantal users don't use g-c-c but use dconf-editor? [14:41] it means we don't get enough people running it yet at this stage to do work in a statistical way :-( [14:41] ev, thanks for the responses [14:41] seb128: anytime :) [14:42] xnox, could be, it means we don't have enough real datas to know what to work on though ... at this point we are basically back to IRC pings and launchpad :p [14:42] seb128: for what it's worth, ubiquity's top problem in 12.10 has 220 instances [14:42] so it could just be that g-c-c is lightly used so far [14:42] (software-center's is 1338) [14:43] * xnox probably should look at those about ubiquity.... [14:43] ev, yeah, could be [14:43] ev, ok, maybe I'm overlooking it but where are retraced stacktraces hidding? === dpm is now known as dpm-laptop [14:43] ev, I would like to get some debug infos for https://errors.ubuntu.com/bucket/?id=failed%3A%2Fusr%2Fbin%2Fgnome-control-center%3A11%3Ax86_64%3A%2Fusr%2Flib%2Fcontrol-center-1%2Fpanels%2Flibnetwork.so%2Bd0e5%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2Bfeca%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B28741%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B29242%3A%2Fus [14:43] r%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2Bfca2%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B20d71%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B29099%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B29242%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibdbus-glib-1.so.2.2.2%2B110a7%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2Bfca2%3A%2Fusr%2Fl [14:43] ib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B20d71%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B29099%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibgobject-2.0.so.0.3200.3%2B29242%3A%2Fusr%2Flib%2Fx86_64-linux-gnu%2Flibdbus-glib-1.so.2.2.2%2B116e6%3A%2Flib%2Fx86_64-linux-gnu%2Flibdbus-1.so.3.5.8%2Be9a6 [14:44] arg [14:44] sorry for the spam, I didn't see that 1 line would do that [14:44] ev, that's the most reported g-c-c bugs for 12.04 in august [14:44] bug [14:47] https://wiki.ubuntu.com/UbuntuDeveloperWeek last day starting in 13 minutes in #ubuntu-classroom [14:48] seb128: that one failed to retrace [14:48] back in a few - need to help a colleague [14:48] ev, ok === rbelem_ is now known as rbelem [15:12] sconklin: that'd be me [15:13] oops [15:13] wrong tab-complete [15:13] scott-work: ^ [15:19] infinity: clang 3.1 ping :) [15:42] Hello [15:42] stgraber: sorry to bother you :) i'll await the ping then === dpm-laptop is now known as dpm === jtechidna is now known as JontheEchidna [16:22] Hello anyone here [16:22] ??? === deryck is now known as deryck[lunch] [16:30] hallyn: re pre-stop screwing up restart.. its bug 703800 [16:30] Launchpad bug 703800 in upstart (Ubuntu) "init: restart command fails to restart main process when pre-stop stanza exists" [Medium,Triaged] https://launchpad.net/bugs/703800 [16:31] SpamapS: yup, thanks, i've marked it a dup [16:39] slangasek: bug 1043936 [16:39] Launchpad bug 1043936 in ubiquity (Ubuntu) "Fresh install of quantal server gets the interface name wrong" [Undecided,New] https://launchpad.net/bugs/1043936 [16:39] slangasek: this is something i mentioned last week in the release channel [16:41] slangasek: i'm just now getting back to it [16:41] bjf: is this with or without preseeding of interface names? [16:41] bjf: fwiw this isn't ubiquity, ubiquity is only used in the desktop live installer [16:41] slangasek: no preseeding of interface names [16:41] ok [16:41] slangasek: i didn't know what to file the bug against [16:42] so it looks like we've failed to get biosdevname included in the target [16:42] bjf: can you confirm that the biosdevname package is missing on the system? [16:42] slangasek: any log files you'd like me to attach? [16:42] slangasek: i had to install biosdevname myself === Quintasan_ is now known as Quintasan [16:43] bjf: have you rebooted since installing biosdevname? [16:43] slangasek: no [16:44] slangasek: can do if you'd like me to [16:44] bjf: so I expect that a reboot will rejigger the interface names now that biosdevname is present, and let things work as intended; we just need to sort out why the package wasn't included in the target [16:45] slangasek: would you like me to reboot and see? [16:45] bjf: if that's ok, yes [16:45] slangasek: np [16:46] slangasek: i have no problem reprovisioning this system from scratch if that would help in any way [16:46] slangasek: it's a test box and only takes about 10 minutes [16:47] bjf: I think if you can attach /var/log/installer/*, and confirm that a reboot is sufficient to fix the networking, that's all we need [16:47] slangasek: just so you know, i had to fix up /etc/network/interfaces by hand to get it onto the network and install biosdevname [16:48] oh, then after reboot your config file and interface names will be criss-crossed ;) [16:48] slangasek: nope, everything came up as eth0 still [16:49] bjf: ok, then /var/log/udev as well please [16:49] ack [16:59] slangasek: logs attached [17:05] Riddell: Did you have any plans to fix the symbols files in telepathy-qt for ARM and PPC? It's been FTBFS all month... [17:23] slangasek: i also attached the preseed file that i'm using [17:30] barry: I don't like OrderedDict. It should support being initialised with dictionary syntax yet preserving order. E.g. OrderedDict({'x':1,'a':1,'b':1}).items() == ['x','a','b'] should be True, but it's not =( [17:30] barry: oh well, will change to tuple now =( [17:32] barry: wait.... i simply don't need a dictionary in this case at all, just tuple of tuples. [17:32] * xnox likes OrderedDicts again [17:33] :) [17:37] xnox: exactly. the former suggestion would be impossible [17:38] barry: phhhh if import print_functions changes print(), you could make former to work !!!! =) [17:39] if barry.is_away(): self.make_black_magic() [17:39] xnox: if you'd like to submit a pep i can commit it for you, but i think it will not live long before the guido hammer fell :) [17:40] barry: maybe on 1st of april [17:40] =) [17:40] yes! we need a new april fools joke :) === deryck[lunch] is now known as deryck === cpg|away is now known as cpg [18:58] slangasek: i added biosdevname to my preseed file pkgsel/include and that seems to have done the trick [18:58] ScottK: developer week session in a few minutes [19:00] bjf: right-o - so we just need to fix it so that's done by default [19:00] slangasek: ack [19:11] \o/ biosdevname [19:20] kees: =)))) if only people new the things that get you excited ! =) [19:21] xnox: it's my secret sysadmin upbringing or something :) [19:22] kees: I'm thinking to get a Linksys WRT router in a plush form now =) [19:22] ah! [19:23] I just built my first dd-wrt router :) [19:55] and if kees drops soon out of IRC we know why :) [20:24] geser: heh, so far I haven't null-routed myself :) === salem_ is now known as _salem [20:52] pitti, around ? [20:53] jbicha: There's a bug in gnomebuntu-default-settings.postrm. It still says Xubuntu. Please fix and reupload. [20:54] i was looking at bug 1018552 . i'm generally python illiterate. [20:54] Launchpad bug 1018552 in apport (Ubuntu) "ubuntu-bug on ec2 causes stack trace" [Undecided,Confirmed] https://launchpad.net/bugs/1018552 [20:54] data/general-hooks/ubuntu.py uses 'startswith' a lot of times. and seems to pass strings in. [20:56] smoser: The issue in that bug is a matter of type changes between python and python3. [20:56] right. [20:57] In python, string and bytes are interchangeable, but in python3 they're different types. [20:57] so i can easily fix that case, with: ami.startswith(b'ami-') [20:57] but the file is loaded with those [20:57] The 'b' at the start makes is of type bytes and not type string. [20:57] so all of thse uses of startswith should be fixed, no? [20:58] It depends on if it really should be bytes and not string. [20:58] Congratulations. IME the is the most painful part of python/python3 porting. [20:59] When you trip over the added distinction in the two types you have to step back and figure out which one you actually want. [20:59] I'd ask barry to look at it as he claims to understand this distinction well. [20:59] ah. is see. [20:59] the url.open returns bytes [20:59] easiest to just change that to a string i think [21:00] Probably, but I haven't looked at the code, so I don't have a very valid opinion. [21:00] smoser: you need to decode/encode it as some kind of string if the url returns a string === sebdebug is now known as seb128 [21:00] sorry, i'm deep into a few other things atm. i could possibly help later [21:02] * ScottK looks for a minion to ping barry every 5 minutes IOT ensure his productivity is completely destroyed. [21:03] ScottK: thanks [21:03] * barry needs no extra help destroying his productivity [21:09] barry, bdrung anyeone.. if you want to look at https://code.launchpad.net/~smoser/ubuntu/quantal/apport/lp1018552/+merge/122147 [21:09] i'm pretty sure that is right (at least it seems to test right) [21:09] but it seems like theres other hplaces there in ubuntu.py that should also need to be fixed. [21:09] smoser: is this python3? [21:10] yes. [21:10] apport is python3 [21:12] smoser: i'm uncomfortable with the blanket str() conversion. probably better to .decode('utf-8') assuming the bytes are coming back as utf-8. the problem is you need to know the encoding, or be told it out of band. or you can just cross your fingers which is what this is pretty much doing ;) [21:12] barry, i changed it [21:12] you're looking at old diff i think [21:12] smoser: ah, sorry [21:12] if ami and ami.startswith(b'ami'): [21:13] yeah, that's probably better [21:13] bytes is bytes :) [21:13] do you think there are othe rthing sin that file that need to be fixed? [21:14] well. i'm gonna upload this one. [21:15] smoser: i can't really tell from the diff context. i'd have to review the code, but i don't have any time right now to do that :( [21:16] fine. [21:16] i'm uploading this one now [21:30] hallyn: Oh, ok. Interesting. That's good to know. Thanks! [21:30] hallyn: (re: upstart jobs and pre-stop and post-stop stanzas) === cpg is now known as cpg|away